Power Credit Report Generation Method, Device, Electronic Device and Readable Storage Medium

Abstract

The present invention provides a method, apparatus, electronic device and readable storage medium for generating an electric power credit report. The method includes: calculating the electric power credit evaluation index value of an enterprise electric power user based on the electricity consumption data information generated by the enterprise electric power user in multiple data centers; analyzing the electricity consumption credit of the enterprise electric power user by means of subjective evaluation plus objective evaluation based on the electric power credit evaluation index value, and generating an electric power credit report for the enterprise electric power user according to a preset template based on the analysis result. By constructing an effective electric power credit evaluation system, the present invention automatically comprehensively analyzes the electric power credit of multiple data centers and generates a credit report by combining subjective and objective evaluations, which can effectively improve the information acquisition efficiency and more intuitively display the true electricity consumption credit situation of electric power users, so as to better serve various credit-related services of enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular, to a method, device, electronic device and readable storage medium for generating an electric power credit report. Background Art

[0002] Modern enterprise credit is not only applied to credit credit, but also widely applied to many aspects such as quality credit, service moral credit, information credit, and health and environmental protection credit in production and operation. Enterprise electricity consumption data has the advantage of reflecting production and operation capabilities and potential. However, currently, when handling business related to user electricity credit, it is necessary to obtain credit information from different relevant data platforms separately and then conduct a comprehensive evaluation, which has strong repeatability and low efficiency. Summary of the Invention

[0003] The present invention provides a method, device, electronic device and readable storage medium for generating an electric power credit report, so as to solve the defect of low efficiency in the prior art and achieve the goal of effectively improving efficiency.

[0004] The present invention provides a method for generating an electric power credit report, including:

[0005] Calculating an electric power credit evaluation index value of the enterprise electric power user based on the electricity consumption data information generated by the enterprise electric power user in multiple data centers;

[0006] Based on the electric power credit evaluation index value, analyzing the electricity credit of the enterprise electric power user by using a subjective evaluation plus objective evaluation method, and based on the analysis result, generating an electric power credit report for the enterprise electric power user according to a preset template.

[0007] The present invention further provides an electric power credit report generating device, including:

[0008] A calculation module, configured to calculate an electric power credit evaluation index value of the enterprise electric power user based on the electricity consumption data information generated by the enterprise electric power user in multiple data centers;

[0009] A report generation module, configured to analyze the electricity credit of the enterprise electric power user by using a subjective evaluation plus objective evaluation method based on the electric power credit evaluation index value, and generate an electric power credit report for the enterprise electric power user according to a preset template based on the analysis result.

[0010] The present invention further provides an electronic device, including a memory, a processor, and a program or instruction stored on the memory and executable on the processor. When the processor executes the program or instruction, the steps of the method for generating an electric power credit report as described in any one of the above are implemented.

[0011] The present invention also provides a non-transitory computer-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a computer, the steps of the power credit report generation method described in any one of the above are implemented.

[0012] The power credit report generation method, device, electronic device and readable storage medium provided by the present invention can effectively improve the information acquisition efficiency and more intuitively display the true electricity consumption credit situation of power users by constructing an effective power credit evaluation system and adopting a combination of subjective and objective evaluations to automatically comprehensively analyze the power credit of multiple data centers and generate a credit report, so as to better serve various credit-related services of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments of the present invention or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0014] Figure 1 It is a schematic flow chart of the power credit report generation method provided by the present invention;

[0015] Figure 2 It is a schematic diagram of the composition of the data analysis basis set in the power credit report generation method provided by the present invention;

[0016] Figure 3 It is a schematic diagram of constructing the overall technical framework of enterprise power credit investigation big data analysis in the power credit report generation method provided by the present invention;

[0017] Figure 4 It is a schematic diagram of the structure of the hierarchical big data analysis warehouse in the power credit report generation method provided by the present invention;

[0018] Figure 5 It is a schematic diagram of the access, calculation and storage process of credit investigation big data based on the data center in the power credit report generation method provided by the present invention;

[0019] Figure 6 It is a schematic diagram of the structure of the enterprise credit investigation big data application integration architecture in the power credit report generation method provided by the present invention;

[0020] Figure 7 It is a schematic diagram of the index with information value greater than 0.1 in the power credit report generation method provided by the present invention;

[0021] Figure 8Schematic diagram of electricity consumption trend waveform in the electricity credit report generation method provided by the present invention;

[0022] Figure 9 Schematic diagram of electricity consumption fluctuation curve in the electricity credit report generation method provided by the present invention;

[0023] Figure 10 Schematic structural diagram of the electricity credit report generation device provided by the present invention;

[0024] Figure 11 Schematic physical structure diagram of the electronic device provided by the present invention. Detailed implementation manners

[0025] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0026] Aiming at the problems of low efficiency in the prior art, etc., the present invention constructs an effective electricity credit evaluation system, adopts a combination of subjective and objective evaluations, automatically comprehensively analyzes the electricity credits of multiple data centers and generates a credit report, which can effectively improve the information acquisition efficiency and can more intuitively display the real electricity consumption credit situation of electricity users, so as to better serve various credit-related services of enterprises. The following will specifically describe and introduce the present invention through multiple embodiments with reference to the accompanying drawings.

[0027] Figure 1 Schematic flowchart of the electricity credit report generation method provided by the present invention, as Figure 1 shown, the method includes:

[0028] S101, calculating the electricity credit evaluation index values of the enterprise electricity users based on the electricity consumption data information generated by the enterprise electricity users in multiple data centers.

[0029] It can be understood that the present invention uses Internet and big data technologies to explore the value of electricity data, that is, for the enterprise electricity users to be analyzed, their electricity consumption data information related to multiple data centers is obtained through the Internet and big data. Then, according to the obtained electricity consumption data information, the corresponding values of a plurality of pre-determined electricity credit evaluation indexes are calculated, that is, the electricity credit evaluation index values.

[0030] Among them, the data middle platform is the precipitation of the business and data of each business unit, constructs a data construction, management and use system including data technology, data governance, data operation, etc., and realizes data empowerment. In the present invention, the data middle platform is a data platform related to the electricity consumption information of power users.

[0031] S102, based on the power credit assessment index value, adopt the method of subjective evaluation plus objective evaluation to analyze the electricity consumption credit of the enterprise power user, and based on the analysis result, generate a power credit report for the enterprise power user according to a preset template.

[0032] It can be understood that, on the basis of determining the power credit assessment index value of the enterprise power user, analyze and determine the high or low, good or bad of the electricity consumption credit of the enterprise power user based on this power credit assessment index value. For example, the good or bad situation of the enterprise's electricity consumption credit can be analyzed from dimensions such as the user's electricity consumption payment situation, business ability evaluation, and / or development potential evaluation. Then, according to the analysis result, generate a corresponding report document to visually display the analysis result. The report document therein can be called a power credit report.

[0033] The power credit report generation method provided by the present invention, by constructing an effective power credit evaluation system, adopting the combination of subjective and objective evaluations, automatically comprehensively analyzes the power credit of multiple data middle platforms and generates a credit report, which can effectively improve the information acquisition efficiency and can more intuitively display the true electricity consumption credit situation of power users, so as to better serve various credit-related businesses of enterprises.

[0034] Furthermore, the power credit report generation method of the present invention further includes at least one of the following operations: constructing an enterprise power credit big data analysis warehouse, and based on the enterprise power credit big data analysis warehouse, obtaining the electricity consumption data information of the enterprise power user; constructing an enterprise power credit comprehensive evaluation index system; constructing an enterprise credit assessment analysis model based on subjective evaluation and objective evaluation.

[0035] It can be understood that before analyzing the electricity consumption data information of the enterprise power user, the present invention can also set and construct one or more of the software and hardware basis for obtaining the electricity consumption data information of the enterprise power user, the basis for evaluation, and the analysis model used for analysis.

[0036] Specifically, when constructing the software and hardware basis of the electricity consumption data information, it can be realized by constructing an enterprise power credit big data analysis warehouse, and in terms of the evaluation basis, the intermediate variables for evaluation can be determined by constructing an enterprise power credit comprehensive evaluation index system. In addition, the analysis model used for analysis can be an enterprise credit assessment analysis model based on subjective evaluation and objective evaluation.

[0037] Specifically, such asFigure 2 As shown in Figure 2 , it is a schematic diagram of the composition of the data analysis basis set in the power credit report generation method provided by the present invention, including: an enterprise power credit investigation big data analysis warehouse, an enterprise power credit investigation comprehensive evaluation index system, and an enterprise credit evaluation model.

[0038] When constructing the enterprise power credit investigation big data analysis warehouse of the present invention, big data technology is applied to enterprise customer data such as internal customer basic information, electricity consumption information, payment information, electricity theft behavior information, power outage and restoration information, business expansion and change information, and user load information, and associated data such as externally set regional gross domestic product and meteorology are integrated to construct a multi-source enterprise electricity consumption credit evaluation analysis library that integrates internal and external data, as shown in Figure 2 part ① in Figure 2 .

[0039] When constructing the enterprise power credit investigation comprehensive evaluation index system, the credit 5C analysis method is adopted. The advantages and values of power data are mined by comparing and contrasting from five aspects: moral character (Character), repayment ability (Capacity), capital strength (Capital), collateral (Collateral), and operating environment conditions (Condition). From five aspects of basic attributes, electricity consumption payment, operating ability, development potential, and electricity regulations, internal and external data are integrated, and multiple evaluation indicators are innovatively planned and designed to scientifically construct the enterprise power credit investigation comprehensive evaluation index system for the target region, as shown in Figure 2 part ② in Figure 2 .

[0040] Among them, the 5C analysis method is one of the expert analysis methods used for credit risk analysis of customers, mainly focusing on comprehensive qualitative analysis of five aspects of the customer's moral character (Character), repayment ability (Capacity), capital strength (Capital), collateral (Collateral), and operating environment conditions (Condition) to judge the customer's payment willingness and payment ability.

[0041] When constructing the enterprise credit evaluation model, big data technologies such as the AHP hierarchical analysis method and the logistic regression algorithm are preferably used to innovatively construct the enterprise credit evaluation model, scientifically evaluate the enterprise's production and operation status, production and operation ability, development potential, electricity consumption trend, etc. in the industry it belongs to, as shown in Figure 2 part ③ in Figure 2 . It can serve small and medium-sized enterprises, financial institutions, etc., be used to promote the development of inclusive financial services, promote the company to enhance the ability to mine and identify customer value, put forward differentiated service suggestions, enhance the company's market competitiveness and retain customers, etc.

[0042] Among them, for the method for generating an electricity credit report provided in the above embodiments, optionally, the construction of the enterprise electricity credit big data analysis warehouse includes: determining the physical resources of the data middle platform required for enterprise electricity credit big data, determining the relevant components of the data middle platform, and constructing the technical framework support for enterprise electricity credit big data analysis; constructing a hierarchical big data analysis model, and determining the implementation logic for accessing, calculating, and storing credit big data based on the data middle platform; building an enterprise credit big data application integration architecture, and integrating the internal and external relevant electricity consumption data of enterprise electricity users through the application integration framework based on the physical resources, the relevant components, the technical framework support, the big data analysis model, and the implementation logic, so as to construct the enterprise electricity credit big data analysis warehouse.

[0043] It can be understood that when constructing the enterprise electricity credit big data analysis warehouse, the physical resources, relevant components and supports, models, data access, calculation and storage processes, etc. required for enterprise electricity credit big data can be configured and constructed separately. Specifically, it includes the following contents:

[0044] 1) Physical resources of the data middle platform required for enterprise electricity credit big data: The system is deployed on the company's data middle platform. The implementation uses the service interfaces and data storage services provided by the enterprise unified cloud platform. Apply for data middle platform resources to deploy the poseidon platform (microservice platform, providing services such as 4A security, organizational structure, and configuration center) on 4 ECSs respectively, deploy the model application service on 2 ECSs, deploy the SFTP file server on 2 ECSs, use 2 RDSs as the data storage databases for multiple independent applications respectively, use the storage resources of the cloud platform for data storage, and use 1 MaxCompute big data computing service for elastic computing. As shown in Table 1, it is an example table of the physical resources of the data middle platform in the method for generating an electricity credit report according to the present invention.

[0045] Table 1. Example table of the physical resources of the data middle platform in the method for generating an electricity credit report according to the present invention

[0046]

[0047] 2) Relevant components and supports of the application's data middle platform: Such as Figure 3As shown in the figure, it is a schematic diagram of constructing the overall technical framework for enterprise power credit investigation big data analysis in the power credit report generation method provided by the present invention. Specifically, based on core components such as the given network data middle platform, application data bus (DataHub), data replication (DTS), big data computing service (MaxCompute), distributed relational database service (DRDS), distributed columnar database (OTS), stream computing engine (BLINK), and data development management platform (Dataworks), core data computing tasks such as data access and data computing and storage are carried out to construct the overall technical framework for enterprise power credit investigation big data analysis.

[0048] 3) Hierarchical big data model design: Apply big data technology to extract enterprise customer data for two years from 2019 to 2020 from a total of 61 database tables of 11 categories of data, including internal customer basic information, electricity consumption information, payment information, illegal electricity use behavior information, power outage and restoration information, business expansion and change information, user load information, etc., and integrate external associated data such as Beijing regional gross domestic product and meteorology to construct a multi-source enterprise electricity credit evaluation analysis library integrating internal and external data as Figure 4 As shown in the figure, it is a schematic diagram of the structure of the hierarchical big data analysis warehouse in the power credit report generation method provided by the present invention. The present invention uses a star model for dimensional modeling of the data warehouse and divides the model logic into a data source layer (ODS), a data common layer (CDM), and an application data layer (ADS) to construct a hierarchical big data analysis warehouse.

[0049] 4) The process of accessing, computing, and storing credit investigation big data based on the data middle platform is as Figure 5 As shown in the figure, it is a schematic diagram of the process of accessing, computing, and storing credit investigation big data based on the data middle platform in the power credit report generation method provided by the present invention, including the following content:

[0050] a) Structured data full volume: Use ETL (Dataworks DI) to access the full volume of data to the source layer MaxCompute, and after data cleaning, governance, modeling, and analysis, enter the data mart ADS / DRDS layer.

[0051] b) Structured data increment: OGG / DTS collects the increment and accesses the data to the source layer through the DataHub bus. If the increment data needs to be calculated and then accessed to the source layer, it needs to be accessed to Blink for analysis and calculation, and the calculation result is re-injected into the DataHub bus and written to the source layer MaxCompute through the bus.

[0052] c) Unstructured files: Images / videos obtain files through active push by the Agent / business system, write them to OSS using the OSS SDK, and use a custom program to extract and analyze the metadata of unstructured data to output a structured structure. The structured result of the unstructured file completes subsequent warehousing operations through the DataHub bus.

[0053] 5) The enterprise credit investigation big data application integration architecture is as Figure 6 shown in the structural schematic diagram of the enterprise credit investigation big data application integration architecture in the power credit report generation method provided by the present invention. Internal company data is obtained through the integrated data middle platform, external data is accessed through offline import, and the result data of the credit investigation model is uploaded to the file server through the enterprise credit investigation file server. The sharing of credit-related data and reports is achieved through the dedicated line interaction files of cooperative financial institutions.

[0054] Optionally, according to the power credit report generation method provided in the above embodiments, the construction of the comprehensive enterprise power credit investigation evaluation index system includes: adopting the credit 5C analysis method, integrating the internal and external relevant power consumption data of the enterprise power users in terms of basic attributes, electricity consumption payment, operation ability, development potential, and power regulations, and constructing the comprehensive enterprise power credit investigation evaluation index system.

[0055] It can be understood that the present invention adopts the credit 5C analysis method to contrast and excavate the advantages and values of power data from the aspects of moral character (Character), repayment ability (Capacity), capital strength (Capital), collateral (Collateral), and operating environment conditions (Condition). From the five aspects of basic attributes, electricity consumption payment, operation ability, development potential, and power regulations, internal and external data are integrated, and 52 evaluation indexes are innovatively planned and designed to scientifically construct the comprehensive enterprise power credit investigation evaluation index system. As shown in Table 2, it is an example table of the enterprise credit evaluation index system according to the present invention.

[0056] Table 2, Example table of the enterprise credit evaluation index system according to the present invention

[0057]

[0058]

[0059] Optionally, according to the power credit report generation method provided in the above embodiments, the construction of the enterprise credit investigation evaluation analysis model based on subjective evaluation and objective evaluation includes: determining the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation, and constructing the enterprise credit investigation evaluation analysis model based on the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation.

[0060] Among them, the implementation logic for determining the subjective evaluation includes: establishing a hierarchical structure model, and constructing a pairwise comparison matrix based on the hierarchical structure model; for each pairwise comparison matrix, calculating the weight vector and performing a consistency test, and based on the verified weight vector, calculating the combined weight vector and performing a combined consistency test; normalizing the verified combined weight vector to obtain the weight coefficients of each power credit evaluation index value, and calculating the subjective evaluation result of the enterprise power user based on the weight coefficients and the power credit evaluation index value.

[0061] It can be understood that when analyzing and evaluating the user's electricity consumption credit, the present invention adopts an analysis method combining subjective and objective methods, that is, establishing a subjective model or implementation logic and an objective model or implementation logic respectively, and then determining the overall analysis model by integrating the subjective analysis and objective analysis results to achieve the subjective and objective analysis of the user's electricity consumption data. Among them, the subjective analysis and evaluation implementation logic includes the following content:

[0062] a) First, establish a hierarchical structure model. On the basis of deeply analyzing the actual problem, decompose the relevant various factors into several levels from top to bottom according to different attributes. The factors at the same level belong to the factors at the upper level or have an impact on the upper-level factors, and at the same time dominate the factors at the lower level or are affected by the lower-level factors. The top layer is the target layer, and the bottom layer is the object layer.

[0063] b) Second, construct a pairwise comparison matrix. Starting from the second layer of the hierarchical structure model, for the factors at the same level that belong to (or affect) each factor at the upper level, use the pairwise comparison method and the 1-9 comparison scale to construct a pairwise comparison matrix until the bottom layer.

[0064] c) Third, calculate the weight vector and perform a consistency test. For each pairwise comparison matrix, calculate the maximum eigenvalue and the corresponding eigenvector, and use the consistency index, random consistency index, and consistency ratio to perform a consistency test. If the test passes, the eigenvector (after normalization) is the weight vector; if not, the pairwise comparison matrix needs to be reconstructed.

[0065] d) Then, calculate the combined weight vector and perform a combined consistency test. Calculate the combined weight vector of the bottom layer with respect to the target, and perform a combined consistency test on the combined weight vector. If the test passes, the decision can be made according to the result represented by the combined weight vector; otherwise, the model needs to be reconsidered or the pairwise comparison matrix with a larger consistency ratio needs to be reconstructed. Among them, the combined consistency verification method for the combined weight vector in the existing analytic hierarchy process can be used to perform the combined consistency verification on the combined weight vector of the present invention.

[0066] e) Then, determine the power credit index weights. After passing the consistency test, normalize the weight vector to obtain the weights of each index.

[0067] f) Finally, calculate the subjective analysis of power credit. After weighted summation of the power credit indicators, multiply by the credit interval (with a full score of 1000 points) to obtain the preliminary credit score of the electricity-consuming enterprise.

[0068] Among them, the optimization of the default and credit loss qualitative coefficient can be carried out as shown in Table 3, which is an example table for optimizing the default and credit loss qualitative coefficient according to the present invention.

[0069] Table 3, Example Table for Optimizing the Default and Credit Loss Qualitative Coefficient According to the Present Invention

[0070] Number of overdue times Qualitative coefficient 1 0.9 2 0.8 3 0.7 4 0.6 >＝5 0.5

[0071] Optionally, on the basis of the above processing, the average industry credit score can also be statistically calculated, that is, the average value of the credit scores of all electricity-consuming enterprises in the same industry is calculated as the average industry credit score.

[0072] Among them, according to the power credit report generation method provided in the above embodiments, optionally, determine the implementation logic of the objective evaluation, including: determining the optimal grouping of each feature corresponding to the pairwise comparison matrix, and calculating the information value of the feature in the optimal grouping; based on the information value, determining the features that meet the preset conditions among the features as quantitative analysis indicators, and based on the quantitative analysis indicators, determining the core analysis indicators; based on the core analysis indicators, calculating the objective evaluation result of the enterprise power user.

[0073] It can be understood that when the present invention implements the objective analysis and evaluation logic of user electricity consumption data information, it may include the following content:

[0074] a) Screen high-quality customers. Based on the subjective evaluation method, set the credit score range to screen high-quality customers.

[0075] b) Statistically calculate the information value. Automatically perform the optimal grouping of each feature, and calculate the influence degree of the feature on the user credit evaluation - the information value (IV). The information value IV is used to measure the correlation between two categorical variables and one of them is a binary variable. The lower the IV value, the weaker the predictive power of the indicator and the lower the correlation; conversely, it indicates that the indicator has a strong correlation with the result variable.

[0076] Let r be the number of groups of the grouping, p i and q i be the percentages of the target variable y recorded in the first category and the second category in the i-th group respectively, that is:

[0077]

[0078]

[0079] c) Select quantitative analysis indicators. Select features with a medium or higher degree of influence (IV > 0.1) as quantitative analysis indicators. By deeply analyzing and calculating the information value IV of the indicator features, indicators with an information value greater than 0.1 are obtained as Figure 7 shown in the schematic diagram of indicators with an information value greater than 0.1 in the power credit report generation method provided by the present invention. Among them, IV < 0.02 indicates that the indicator feature has no predictive ability; 0.02 <= IV < 0.1 indicates weak predictive ability; 0.1 <= IV < 0.3 indicates medium predictive ability; 0.3 <= IV < 0.5 indicates strong predictive ability; > 0.5 indicates very strong predictive ability, and the indicator feature seriously affects user evaluation. Select features with a medium or higher predictive ability (IV > 0.1) as quantitative analysis indicators. The specific indicators are shown in Table 4 below, which is an example table of quantitative analysis indicators according to the present invention.

[0080] Table 4, Example Table of Quantitative Analysis Indicators According to the Present Invention

[0081]

[0082]

[0083] Among them, indicators with relatively large IV values such as the cumulative overdue days, cumulative overdue times, average payment return duration, current electricity price month-on-month growth rate, current month's electricity bill, current month's electricity consumption, and monthly electricity consumption month-on-month growth rate also basically correspond to the indicators with relatively large weights in the qualitative analysis, indicating that the indicator weights based on expert experience in the qualitative analysis and the indicator weights based on data features in the quantitative analysis reach a high similar level, meeting expectations.

[0084] d) Determine the core indicators. Continuing to use the enterprise credit rating score of 700 in the qualitative analysis as the criterion for good credit, dividing 250,000 enterprise users as samples into 30% and 70%, selecting 70% of the 175,176 users to perform feature training on 30 indicators obtained by feature screening, extracting data features, obtaining 30 types of features with an IV value greater than 0.1 after binning, and predicting the remaining 30% of the 75,076 users to verify the accuracy of feature extraction. The logistic regression formula is as follows:

[0085]

[0086] Among them, x is 30 features after feature screening and binning processing, which is a 1×30-dimensional vector, y is the user identifier with 700 points as the boundary (where y = 1 is a high-credit user and y = 0 is other users), and σ represents the conventional sigmod (i.e., S-shaped growth curve) function.

[0087] Finally, through iterative training, the weight parameter W = [w0, w1, w2, …, w 30 is obtained, and these weight parameters are involved in the calculation of the user's score. As shown in Table 5, it is an example table of the weight parameters of the quantitative analysis index according to the present invention.

[0088] Table 5, Example Table of Weight Parameters of Quantitative Analysis Index According to the Present Invention

[0089]

[0090]

[0091] On the basis of the above processing, the AUC (Area Under Curve) method based on ROC curve analysis is used to evaluate the selected logistic regression model. Through accuracy verification, the AUC of the logistic regression model is obtained as 99.85%, indicating that the influence degree of the selected logistic regression model is relatively strong.

[0092] Among them, when calculating AUC, first sort the probability score values of the samples belonging to the positive samples from large to small, and then let the serial number of the sample corresponding to the largest score in this sorting be n, the serial number of the sample corresponding to the second largest score be n - 1, and so on. Then, add up the serial numbers of all the positive samples, and subtract the constant term M(M + 1) / 2. Finally, divide the result of the above subtraction by M×N, and the AUC is obtained as follows:

[0093]

[0094] Among them, M and N respectively represent the number of positive samples and the number of negative samples, and ∑ i∈pos rank i represents the cumulative sum of the serial numbers of the positive samples, and pos represents the set of positive samples.

[0095] Through accurate verification calculation, the AUC of the logistic regression model is obtained as 99.85%, and the accuracy verification shows that the selected logistic regression model has good prediction ability.

[0096] Among them, AUC is the area under the ROC curve, that is, the area of the region enclosed by the ROC curve and the coordinate axes. Since the ROC curve is generally above the line y = x, the value range is between 0.5 and 1. Using AUC as an evaluation index is because the ROC curve often cannot clearly show which classifier has a better effect, and as a numerical value, the larger the value of AUC, the better the classifier effect.

[0097] The receiver operating characteristic curve (ROC curve) is a graphical analysis tool used for:

[0098] (1) Selecting the best signal detection model and discarding sub-optimal models;

[0099] (2) Setting the optimal threshold within the same model.

[0100] ROC can be used to compare the relevant performance of different classifiers. Among them, the abscissa is the false positive rate (FPR), and the ordinate is the true positive rate (TPR).

[0101] In FPR, it represents how many of all negative examples are predicted as positive examples; in TPR, it represents how many true positive examples are predicted; the closer the ROC curve is to the upper left corner, the better the performance of the classifier, which means that the classifier obtains a high true positive rate while having a very low false positive rate.

[0102] e) Determining the objective credit score based on the core indicators. Applying the credit scoring card method, the formula for the quantitative analysis model is as follows:

[0103]

[0104] BaseScore = q + w0 * p.

[0105] Among them, User_Score represents the objective evaluation credit score of the user, BaseScore represents the objective evaluation basic score of the user, x i represents the core indicator, w i represents the weight coefficient of the core indicator, p represents the scale parameter of the scoring card, q represents the compensation parameter of the scoring card, and w0 represents the basic weight, which is a constant.

[0106] According to the above embodiments, a basic score of 700 can be set, and at the same time, the packet data optimization parameter PDO (representing the change value of the score when the good-bad ratio doubles) can be set to 10 (the good-bad ratio doubles every 10 points), and the good-bad ratio is taken as 10. Then, based on the above-set benchmark score, PDO, and good-bad ratio, p = 10 / log(2) and q = 700 - 10 * log(10) / log(2) can be determined, where w i is taken from the weight vector W obtained by logistic regression.

[0107] Optionally, constructing the enterprise credit assessment and analysis model includes: constructing an operation model for the subjective evaluation result and the objective evaluation result as the enterprise credit assessment and analysis model.

[0108] It can be understood that, based on the above determination of subjective evaluation results and objective evaluation results, a comprehensive credit score can be determined accordingly, that is, weights for the two credit evaluation results are set respectively, and an enterprise credit investigation and evaluation analysis model is determined based on the weighted method, so that the subjective and objective comprehensive credit score can be calculated.

[0109] Among them, according to the power credit report generation method provided in each of the above embodiments, optionally, generating the power credit report for the enterprise power user includes: based on the analysis result, outputting the power credit report including: enterprise basic information and enterprise credit evaluation, and the enterprise credit evaluation includes enterprise credit evaluation criteria and enterprise credit scores.

[0110] It can be understood that, based on the data analysis according to each of the above embodiments, a user electricity consumption credit report can be generated according to the analysis result to visually display the user's electricity consumption credit. Specifically, when outputting the credit report, the following contents can be included:

[0111] First, output the enterprise basic information.

[0112] Enterprise name: XXXX Co., Ltd.

[0113] Power consumption address: Plot XX, Block XX, XXXXXXXXXX Street

[0114] Account opening date: XX / XX / XXXX

[0115] Industry classification: Industry - Manufacturing - Automobile manufacturing

[0116] Contract capacity: 40000 kVA Operating capacity: 40000 kVA

[0117] Secondly, output the enterprise credit evaluation, including:

[0118] 1) Output the evaluation criteria.

[0119] Final credit score = Qualitative analysis credit score * 60% + Quantitative analysis credit score * 40%.

[0120] The total credit score is set to 1000. According to the customer's score, it is divided into five credit levels: A credit - excellent user, B credit - good user, C credit - average user, D credit - poor user, and E credit - extremely - poor user. As shown in Table 6, it is an example table of qualitative analysis credit levels according to the present invention.

[0121] Table 6, Example table of qualitative analysis credit levels according to the present invention

[0122] Serial number Credit rating Scoring range 1 A (Users with excellent credit) [760-1000] 2 B (Users with good credit) [700-760) 3 C (Users with average credit) [500-700) 4 D (Users with poor credit) [400-500) 5 E (Users with extremely poor credit) [0-400)

[0123] a) Qualitative analysis scoring rules: Based on the judgment matrix filled in by experts, analyze the importance of each measurement dimension, obtain the weight of each measurement dimension, and sort out the enterprise credit scoring rules, including: the weight of basic attributes is 3%, the weight of electricity payment is 40%, the weight of operating ability is 30%, the weight of development potential is 17%, and the weight of electricity regulations is 10%. As shown in Table 7, it is an example table of the enterprise credit scoring rules according to the present invention.

[0124] Table 7, Example Table of Enterprise Credit Scoring Rules According to the Present Invention

[0125] Serial number Dimension Weight Total score 1 Basic attributes 3％ 30 2 Electricity payment 40％ 400 3 Operating ability 30％ 300 4 Development potential 17％ 170 5 Electric power regulations 10％ 100

[0126] Use the analytic hierarchy process to calculate the score. Let Score_F be the credit rating score of the customer, and the credit score can be obtained. The calculation formula is as follows:

[0127] Score_F = (S m × w1 + S n × w2 + S p × w3 + S k × w4 + S q × w5) × K.

[0128] Among them, S m 、S n 、S p 、S k 、S q are the credit scores of basic attributes, electricity payment, operating ability, development potential, and electricity regulations respectively; w1, w2, w3, w4, w5 are the weights of basic attributes, electricity payment, operating ability, development potential, and electricity regulations respectively; K is the mapping coefficient of qualitative indicators; Score_F is the final credit score.

[0129] b) Qualitative coefficient of default and dishonesty.

[0130] Table 8 shows an example table of the qualitative coefficient of default and dishonesty according to the present invention.

[0131] Table 8, Example Table of Qualitative Coefficient of Default and Dishonesty According to the Present Invention

[0132] Qualitative indicators Behavior type Qualitative coefficient K Power theft Serious credit violation 0.2 Illegal electricity use General credit violation 0.3 Legal collection General credit violation 0.3

[0133] For example, if the number of arrears or overdue times in the past 12 months exceeds 1 time, it is a minor dishonesty behavior of 0.5 - 0.9.

[0134] c) Quantitative analysis scoring rules.

[0135] The indicators obtained by using the AHP (Analytic Hierarchy Process) form characteristic data, and some indicators with low influence and related indicators (such as average payment collection duration and cumulative overdue days) are discarded. The characteristic data is applied, and a quantitative credit score is calculated using the logistic regression algorithm.

[0136] 2) Output the enterprise credit score.

[0137] Suppose the credit system evaluation score of a certain enterprise in this period is 769 points, and the comprehensive credit rating is A (extremely creditworthy user). The average score of enterprises in the same industry (industry - manufacturing, the same industry is taken for the following intra - industry comparisons) is 623 points. The credit score of this enterprise exceeds 97.7% of the users in the same industry.

[0138] Output the specific sub - item index scores of this enterprise as shown in Table 9, which is an example table of sub - item index scores according to the present invention.

[0139] Table 9, Example Table of Sub - item Index Scores According to the Present Invention

[0140] Qualitative score Basic attributes Electricity payment Electric power regulations Operating ability Development potential 780 703 787 1000 716 726

[0141] Output the qualitative coefficient: Suppose this enterprise has no default electricity consumption and electricity theft behaviors in the past two years, no legal collection and overdue payment behaviors, and the qualitative coefficient is 1.

[0142] Furthermore, on the basis of the above - mentioned embodiments, the present invention can also conduct enterprise evaluation and analysis. The following enterprise evaluation and analysis mainly gives examples from the dimension of qualitative analysis.

[0143] 1. Basic attributes.

[0144] The above - mentioned user's basic attribute score is 703 points, which is 195 points lower than the industry average level of 898 points. Specific analysis: The account age is relatively young, the payment method is bank collection, and it belongs to a non - high - energy - consuming industry, so the basic attributes are good.

[0145] 2. Electricity consumption and payment This user's electricity consumption and payment score is 787 points, which is 65 points higher than the industry average level of 722 points.

[0146] 1) Payment behavior: Payment behavior measures the user's performance ability and performance effect. The total electricity bills issued in the past two years are 24 times, and there is no overdue electricity bill payment situation (the industry average overdue payment rate is 5%). The enterprise's average payment collection duration (collection time - issue date) is 3.4 days, which is 6.3 days shorter than the industry average payment duration of 9.7 days, indicating active payment.

[0147] 2) Payment method: The payment method can reflect the enterprise's disposable property status. This enterprise is a user with installment transfer. In the past two years, this user has made a total of 61 payments (including installment transfer payments), and all payments are made through bank collection. The preference for the payment method is excellent.

[0148] 3. Electric power regulations.

[0149] The user's score for electric power regulations is 1000 points, 1 point higher than the industry average of 999 points. In the past two years, the enterprise has had no default electricity consumption or electricity theft behavior.

[0150] 4. Operating ability.

[0151] The user's score for operating ability is 716 points, 58 points higher than the industry average of 658 points.

[0152] 1) Electricity consumption scale: The electricity consumption situation represents the energy consumption situation of the enterprise. The higher this indicator, the stronger the energy consumption of the enterprise. The average monthly electricity consumption of the enterprise in the past two years: 9,203,975 (kWh), which is 174 times the average electricity consumption of the industry (53,084 kWh), indicating a large electricity consumption scale.

[0153] 2) Enterprise electricity bill: The enterprise electricity bill directly reflects the energy value of the enterprise and objectively reflects the business situation of the enterprise. As Figure 8 shown, it is a schematic diagram of the electricity bill trend waveform in the method for generating an electric power credit report provided by the present invention. In the past two years, the cumulative electricity bill payable by the enterprise is 173.973 million yuan, and the actual payment is 173.973 million yuan, which is 122.8 times the average cumulative electricity bill of the industry of 1.417 million yuan.

[0154] 3) Electricity capacity: The operating capacity of the enterprise is 40000 kVA, and the average operating capacity of the industry is 542 KW. The operating capacity of the enterprise is 73.8 times the industry average level.

[0155] 5. Development potential.

[0156] The user's score for operating ability is 726 points, 64 points higher than the industry average of 658 points.

[0157] 1) Electricity consumption trend: It represents the electricity consumption potential of the enterprise. The higher this indicator, the stronger the development momentum of the enterprise's energy consumption. In the past two years, the cumulative electricity consumption of this user is 2.21 billion kWh, and the average cumulative electricity consumption of the industry is 1.63 million kWh. The electricity consumption potential of the enterprise is higher than the industry average level. The growth rate of electricity consumption in the second half of 2019 compared with 2018 is 3.6%, and the decrease rate in the first half of 2020 compared with the same period last year is 7.1%.

[0158] 2) Electricity consumption fluctuation level: The electricity consumption fluctuation level measures the stability of the enterprise's electricity consumption. The lower this indicator, the more stable the production and operation of the enterprise. As Figure 9 shown, it is a schematic diagram of the electricity consumption fluctuation curve in the method for generating an electric power credit report provided by the present invention. In the past 12 months, the average monthly volatility of the enterprise is 12.1%, and the average volatility of the industry is 18.4%. The electricity consumption fluctuation level is 34% lower than the industry average.

[0159] 3) Trends of capacity increase and decrease: There has been no capacity increase or decrease business in the past two years.

[0160] Based on the same inventive concept, the present invention further provides a power credit report generation device according to the above embodiments. This device is used to implement the generation of power credit reports in the above embodiments. Therefore, the descriptions and definitions in the power credit report generation methods of the above embodiments can be used for the understanding of each execution module in the present invention. For details, reference can be made to the above method embodiments and will not be elaborated here.

[0161] According to an embodiment of the present invention, the structure of the power credit report generation device is as Figure 10 shown, which is a schematic structural diagram of the power credit report generation device provided by the present invention. This device can be used to implement the generation of power credit reports in the above method embodiments. The device includes: a calculation module 1001 and a report generation module 1002. Among them:

[0162] The calculation module 1001 is used to calculate the power credit evaluation index value of the enterprise power user based on the power consumption data information generated by the enterprise power user in multiple data centers; the report generation module 1002 is used to analyze the power consumption credit of the enterprise power user by combining subjective evaluation and objective evaluation based on the power credit evaluation index value, and generate a power credit report for the enterprise power user according to a preset template based on the analysis result.

[0163] The power credit report generation device provided by the present invention can effectively improve the information acquisition efficiency by constructing an effective power credit evaluation system and adopting a combination of subjective and objective evaluations, and can more intuitively display the true power consumption credit situation of power users, so as to better serve various credit-related businesses of enterprises.

[0164] Optionally, the power credit report generation device of the present invention further includes a construction module, which is used to perform at least one of the following operations:

[0165] Construct a big data analysis warehouse for enterprise power credit investigation, and obtain the power consumption data information of the enterprise power user based on the big data analysis warehouse for enterprise power credit investigation;

[0166] Construct an integrated evaluation index system for enterprise power credit investigation;

[0167] Construct an enterprise credit investigation evaluation analysis model based on subjective evaluation and objective evaluation.

[0168] Optionally, when the construction module is used to construct the big data analysis warehouse for enterprise power credit investigation, it is used for:

[0169] Determine the physical resources of the data center required for the enterprise power credit big data, and determine the relevant components of the data center, and build a technical framework support for the enterprise power credit big data analysis;

[0170] Build a layered big data analysis model and determine the credit investigation big data access and computing storage implementation logic based on the data middle platform;

[0171] Build an enterprise credit big data application integration architecture, and based on the physical resources, the relevant components, the technical framework support, the big data analysis model and the implementation logic, integrate the internal and external relevant electricity consumption data of the enterprise power users through the application integration framework to build the enterprise power credit big data analysis warehouse.

[0172] Optionally, the construction module, when used for constructing the enterprise power credit comprehensive evaluation index system, is used to:

[0173] By adopting the credit 5C analysis method and integrating the internal and external relevant electricity consumption data of the corporate electricity users in terms of basic attributes, electricity payment, operating ability, development potential and electricity regulations, a comprehensive evaluation index system for the corporate electricity credit is constructed.

[0174] Optionally, the construction module, when used for constructing the enterprise credit assessment analysis model based on subjective evaluation and objective evaluation, is used to:

[0175] Determining the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation, and constructing the enterprise credit assessment analysis model based on the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation;

[0176] Wherein, determining the implementation logic of the subjective evaluation includes:

[0177] Establishing a hierarchical model, and constructing a pairwise comparison matrix based on the hierarchical model;

[0178] For each of the pairwise comparison matrices, a weight vector is calculated and a consistency check is performed, and based on the verified weight vector, a combined weight vector is calculated and a combined consistency check is performed;

[0179] The combined weight vector after normalization processing is verified to obtain the weight coefficient of each power credit assessment index value, and based on the weight coefficient and the power credit assessment index value, the subjective evaluation result of the enterprise power user is calculated.

[0180] Optionally, the construction module, when used to determine the implementation logic of the objective evaluation, is used to:

[0181] Determine the optimal grouping of each feature corresponding to the pairwise comparison matrix, and calculate the information value of the features in the optimal grouping;

[0182] Based on the information value, determine the features that meet the preset conditions among the features as quantitative analysis indicators, and based on the quantitative analysis indicators, determine the core analysis indicators;

[0183] Based on the core analysis indicators, calculate the objective evaluation result of the enterprise power users;

[0184] The construction of the enterprise credit assessment analysis model includes:

[0185] Construct an operation model for the subjective evaluation result and the objective evaluation result as the enterprise credit assessment analysis model.

[0186] Optionally, when the report generation module is used to generate the power credit report for the enterprise power users, it is used for:

[0187] Based on the results of the analysis, output the power credit report including: enterprise basic information and enterprise credit evaluation, and the enterprise credit evaluation includes enterprise credit evaluation criteria and enterprise credit scores.

[0188] It can be understood that in the present invention, the above-mentioned relevant program modules in the devices of the above-mentioned embodiments can be implemented by a hardware processor (hardware processor). And, the power credit report generation device of the present invention utilizes the above-mentioned program modules to be able to implement the power credit report generation process of the above-mentioned method embodiments. When used to implement the power credit report generation in the above-mentioned method embodiments, the beneficial effects produced by the device of the present invention are the same as those of the corresponding above-mentioned method embodiments, and reference can be made to the above-mentioned method embodiments, which will not be elaborated here.

[0189] As another aspect of the present invention, in this embodiment, according to the above-mentioned embodiments, an electronic device is provided. The electronic device includes a memory, a processor, and a program or instruction stored on the memory and executable on the processor. When the processor executes the program or instruction, the steps of the power credit report generation method as described in the above-mentioned embodiments are implemented.

[0190] Furthermore, the electronic device of the present invention may further include a communication interface and a bus. Refer to Figure 11 , the entity structure diagram of the electronic device provided by the present invention includes: at least one memory 1101, at least one processor 1102, a communication interface 1103, and a bus 1104.

[0191] Among them, the memory 1101, the processor 1102, and the communication interface 1103 complete mutual communication through the bus 1104. The communication interface 1103 is used for information transmission between the electronic device and the big data device; the memory 1101 stores programs or instructions that can run on the processor 1102. When the processor 1102 executes the programs or instructions, the steps of the power credit report generation method described in the above embodiments are implemented.

[0192] It can be understood that the electronic device at least includes a memory 1101, a processor 1102, a communication interface 1103, and a bus 1104. The memory 1101, the processor 1102, and the communication interface 1103 form a mutual communication connection through the bus 1104 and can complete mutual communication. For example, the processor 1102 reads program instructions of the power credit report generation method from the memory 1101. In addition, the communication interface 1103 can also implement a communication connection between the electronic device and the big data device and can complete mutual information transmission. For example, the power consumption data information of enterprise power users can be read through the communication interface 1103.

[0193] When the electronic device runs, the processor 1102 calls the program instructions in the memory 1101 to execute the methods provided in the above method embodiments. For example, it includes: calculating the power credit evaluation index value of the enterprise power user based on the power consumption data information of the enterprise power user; analyzing the power consumption credit of the enterprise power user by using a subjective evaluation plus an objective evaluation method based on the power credit evaluation index value, and generating a power credit report for the enterprise power user based on the analysis result.

[0194] When the program instructions in the above-mentioned memory 1101 can be implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Alternatively, all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0195] The present invention also provides a non-transitory computer-readable storage medium according to the above embodiments, on which programs or instructions are stored. When the programs or instructions are executed by a computer, the steps of the power credit report generation method described in the above embodiments are implemented.

[0196] As another aspect of the present invention, this embodiment further provides a computer program product according to the above embodiments. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the power credit report generation method provided by the above method embodiments.

[0197] The electronic device, non-transitory computer-readable storage medium, and computer program product provided by the present invention, by executing the steps of the power credit report generation method described in the above embodiments, by constructing an effective power credit evaluation system, adopting a combination of subjective and objective evaluations, automatically comprehensively analyzing the power credit of multiple data centers and generating a credit report, can effectively improve the information acquisition efficiency, and can more intuitively display the true electricity consumption credit situation of power users, so as to better serve various credit-related businesses of the enterprise.

[0198] It can be understood that the embodiments of the device, electronic device, and storage medium described above are merely illustrative. The units described as separate components may or may not be physically separated, and may be located in one place or distributed to different network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0199] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or equivalently replace some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating an electricity credit report, characterized in that, Including: Calculating the power credit assessment index value of the enterprise power user based on the power consumption data information generated by the enterprise power user in multiple data centers; Based on the power credit assessment index value, analyzing the power consumption credit of the enterprise power user by means of subjective evaluation plus objective evaluation, and generating a power credit report for the enterprise power user according to a preset template based on the analysis result; It also includes: constructing a big data analysis warehouse for enterprise power credit investigation; the construction of the big data analysis warehouse for enterprise power credit investigation includes: Determining the physical resources of the data center required for the big data of enterprise power credit investigation, and determining the relevant components of the data center to construct the technical framework support for the big data analysis of enterprise power credit investigation; Constructing a hierarchical big data analysis model, and determining the implementation logic for accessing, calculating and storing the credit investigation big data based on the data center; Building an application integration architecture for enterprise credit investigation big data, and integrating the internal and external relevant power consumption data of the enterprise power user through the application integration architecture based on the physical resources, the relevant components, the technical framework support, the big data analysis model and the implementation logic to construct the big data analysis warehouse for enterprise power credit investigation; It also includes: constructing an enterprise credit investigation assessment analysis model based on subjective evaluation and objective evaluation; The construction of the enterprise credit investigation assessment analysis model based on subjective evaluation and objective evaluation includes: Determining the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation, and constructing the enterprise credit investigation assessment analysis model based on the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation; Among them, determining the implementation logic of the subjective evaluation includes: Establishing a hierarchical structure model, and constructing a pairwise comparison matrix based on the hierarchical structure model; For each of the pairwise comparison matrices, calculating the weight vector and performing a consistency test, and calculating the combined weight vector and performing a combined consistency test based on the verified weight vector; Normalizing the verified combined weight vector to obtain the weight coefficients of each power credit assessment index value, and calculating the subjective evaluation result of the enterprise power user based on the weight coefficients and the power credit assessment index value; Determining the implementation logic of the objective evaluation includes: Determining the optimal grouping of each feature corresponding to the pairwise comparison matrix, and calculating the information value of the feature in the optimal grouping; Based on the information value, determining the features that meet the preset conditions in the features as quantitative analysis indicators, and determining the core analysis indicators based on the quantitative analysis indicators; Calculating the objective evaluation result of the enterprise power user based on the core analysis indicators; Among them, the preset condition includes: the corresponding information value is greater than 0.1; The calculation of the objective evaluation result of the enterprise power user based on the core analysis indicators includes: Determining the objective evaluation credit score of the enterprise power user based on the following formula: ; ; Among them, User_Score represents the objective evaluation credit score of the user, BaseScore represents the objective evaluation base score of the user, x i represents the core analysis index, w i represents the weight coefficient of the core analysis index, p represents the scale parameter of the scoring card, q represents the compensation parameter of the scoring card, w 0 represents the basic weight, which is a constant; The construction of the enterprise credit investigation assessment analysis model includes: Constructing an operation model for the subjective evaluation result and the objective evaluation result as the enterprise credit investigation assessment analysis model.

2. The method for generating an electricity credit report according to claim 1, wherein It also includes at least one of the following operations: Based on the enterprise power credit investigation big data analysis warehouse, obtain the electricity consumption data information of the enterprise power users; Construct an integrated evaluation index system for enterprise power credit investigation.

3. The method for generating an electricity credit report according to claim 2, wherein The construction of the integrated evaluation index system for enterprise power credit investigation includes: Adopt the credit 5C analysis method, integrate the internal and external relevant electricity consumption data of the enterprise power users in terms of basic attributes, electricity consumption payment, operation ability, development potential and power regulations, and construct the integrated evaluation index system for enterprise power credit investigation.

4. The method for generating an electricity credit report according to any one of claims 1 to 3, characterized in that, The generation of the power credit report for the enterprise power users includes: Based on the analysis results, the output of the power credit report includes: enterprise basic information and enterprise credit evaluation, and the enterprise credit evaluation includes enterprise credit evaluation criteria and enterprise credit scores.

5. An electric power credit report generation device, characterized in that It includes: A calculation module, configured to calculate the power credit investigation evaluation index value of the enterprise power user based on the electricity consumption data information generated by the enterprise power user in multiple data centers; A report generation module, configured to analyze the electricity consumption credit of the enterprise power user based on the power credit investigation evaluation index value by means of subjective evaluation plus objective evaluation, and generate a power credit report for the enterprise power user according to a preset template based on the analysis results; The power credit report generation device further includes a construction module, configured to perform at least one of the following operations: Construct an enterprise power credit investigation big data analysis warehouse; Construct an enterprise credit investigation evaluation analysis model based on subjective evaluation and objective evaluation; When the construction module is used for constructing the enterprise power credit investigation big data analysis warehouse, it is used for: Determine the physical resources of the data center required for enterprise power credit investigation big data, and determine the relevant components of the data center, and construct the technical framework support for enterprise power credit investigation big data analysis; Construct a hierarchical big data analysis model, and determine the implementation logic of credit investigation big data access, calculation and storage based on the data center; Build an enterprise credit investigation big data application integration architecture, and integrate the internal and external relevant electricity consumption data of the enterprise power users through the application integration architecture based on the physical resources, the relevant components, the technical framework support, the big data analysis model and the implementation logic, and construct the enterprise power credit investigation big data analysis warehouse; When the construction module is used for constructing the enterprise credit investigation evaluation analysis model based on subjective evaluation and objective evaluation, it is used for: Determine the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation, and construct the enterprise credit investigation evaluation analysis model based on the implementation logic of the subjective evaluation and the implementation logic of the objective evaluation; Among them, determining the implementation logic of the subjective evaluation includes: Establish a hierarchical structure model, and construct a pairwise comparison matrix based on the hierarchical structure model; For each of the pairwise comparison matrices, calculate the weight vector and perform a consistency test, and calculate the combined weight vector and perform a combined consistency test based on the verified weight vector; Normalize the combined weight vector after verification to obtain the weight coefficients of each of the power credit evaluation index values, and calculate the subjective evaluation result of the enterprise power user based on the weight coefficients and the power credit evaluation index values; When determining the implementation logic for the objective evaluation, the construction module is configured to: Determine the optimal grouping of each feature corresponding to the pairwise comparison matrix, and calculate the information value of the features in the optimal grouping; Based on the information value, determine the features that meet the preset conditions among the features as quantitative analysis indicators, and determine the core analysis indicators based on the quantitative analysis indicators; Calculate the objective evaluation result of the enterprise power user based on the core analysis indicators; wherein, the preset condition includes: the corresponding information value is greater than 0.1; Calculating the objective evaluation result of the enterprise power user based on the core analysis indicators includes: Determine the objective evaluation credit score of the enterprise power user based on the following formula: ; ; Among them, User_Score represents the objective evaluation credit score of the user, BaseScore represents the objective evaluation base score of the user, x i represents the core analysis index, w i represents the weight coefficient of the core analysis index, p represents the scale parameter of the scoring card, q represents the compensation parameter of the scoring card, w 0 represents the base weight, which is a constant; Constructing the enterprise credit evaluation analysis model includes: Construct an operation model for the subjective evaluation result and the objective evaluation result as the enterprise credit evaluation analysis model.

6. An electronic device, comprising a memory, a processor, and a program or instruction stored on the memory and executable on the processor, characterized in that, When the processor executes the program or instruction, the steps of the power credit report generation method according to any one of claims 1 to 4 are implemented.

7. A non-transitory computer-readable storage medium having a program or instructions stored thereon, characterized in that, When the program or instruction is executed by a computer, the steps of the power credit report generation method according to any one of claims 1 to 4 are implemented.

Images